Dietary choices, and in particular, excess calorie intake leading to obesity, are strong, but reversible risk factors for cancer. For example, foods high in solid fats and added sugars (SoFaS) are low-nutrient, high calorie foods that increase the risk of cancer by promoting weight gain. As such, the reduction of SoFaS is consistent with American Institute for Cancer Research and the American Cancer Society dietary recommendations. Behavioral interventions to alter diet have limited long-term efficacy, most likely because eating decisions are governed by automatic neurocognitive processes that are not addressed in conventional interventions. In particular, the ability to refrain from consuming unhealthy, but widely available, palatable foods, is increasingly understood to depend on inhibitory control, i.e., the ability to cut off action tendencies that are put in motion by innate drives towards rewarding behaviors. Recent work by our team and others have demonstrated that computer-based inhibitory control trainings result in short-term, specific changes in behavior, such as reducing intake of salty snack food, chocolate, and alcoholic beverages. An automatized, home computer-based inhibitory control training offers the potential of an inexpensive and highly disseminable method of lowering cancer risk across wide swaths of the population. As such, we aim to evaluate the feasibility, acceptability, mechanism of action, effectiveness and persistence of a home computer-based inhibitory control training. In particular, we hypothesize that a high-repetition training in inhibitory control will result in increased adherence to a low-SoFaS diet, and that effects will be mediated through improved inhibitory control. We further hypothesize the training will be most effective for those starting of with impaired inhibitory control, as well as those with strongest desire for palatable foods and those with strongest explicit health goals. Lastly, we aim to examine the impact of inhibitory control training on secondary outcomes, including on overall caloric intake, and on short-term weight loss. To achieve these aims, the proposed study will recruit 150 overweight and obese individuals who currently eat high-SoFaS diets, and who wish to improve their diets. Participants will be assigned a reduced-SoFaS diet for 12 weeks. After a baseline period, participants will be randomized to receive 6 weeks of either inhibitory control training or a sham training. The 6-week intervention will consist of 15 minutes per day of home computer- based inhibitory control training, and will be followed by a 2-week booster and then 2-week follow-up period. Dietary adherence will be measured via a customized smartphone app that will prompt repeated recording of targeted food consumption (i.e., ecological momentary assessment; EMA) and via automated 24-hour food recall. Neurocognitive variables will be assessed pre and post-training in order to test trainings' mechanism of action, and moderation will be assessed through baseline trait measures of explicit health goals, implicit attitudes towards appetitive stimuli, boy mass index, and responsivity to food cues.